We have developed a sensitive ELISA assay for measuring the potency of trivalent poliovirus vaccines using mouse monoclonal antibodies for antigen detection. During routine tests of IPV we found a vaccine lot the type 2 component of which tested very low with our standard reagent; but gave satisfactory values with another type 2 monoclonal antibody or rabbit polyclonal sera. We are collaborating with another laboratory as a separate research project to identify the site specificity of our monoclonal antibodies. We are in the process of trying to elucidate the cause for this apparent epitope specific change. We reviewed the literature for chemicals known to affect the potency of poliovirus. We examined the effect of thimerosal on the potency of IPV. Preparations were held at three temperatures 37, 25 and 4 degrees C for 2 weeks. The preparations were then tested for potency by ELISA at 2, 10, and 14 days. Vaccine preparations held at 37 and 4 degrees degrees C were inoculated into mice on day 14. We found that combinations of IPV and thimerosal held for 14 days resulted in a loss of potency when kept at 25 degrees C as well as 37 degrees C for poliovirus types 1 and 2. Poliovirus type 3 was less sensitive to thimerosal and lost potency when the vaccine was held at 37 degrees C only. This suggested that handling of vaccines with thimerosal is very important in terms of IPV potency. We observed that our standard type 2 monoclonal antibody appeared to be directed against an epitope that had undergone changes in the presence of thimerosal. The potency of the vaccine measured in vitro did not correlate with immunogenicity in mice. These findings have important implications for manufacturers of combination vaccines as they will need to consider the effect of each preservative on the antigens in the combination.